EFDA-JET-CP(03)01/65

Disruption Classification at JET with Neural Techniques

Neural networks have been trained to classify different types of plasma disruptions in a Tokamak experiment using several diagnostic signals as input. Tests refer to data collected from disrupted pulses that occurred during four years of JET experiments. The results show the feasibility of a reliable neural network classifier.
Name Size  
EFDC030165 827.24 Kb